The contribution of congenital anomalies to infant mortality

Transcription

1 Inequalities in Infant Mortality Project Briefing Paper 4 The contribution of congenital anomalies to infant mortality Jennifer J Kurinczuk, Jennifer Hollowell, Patricia A Boyd, Laura Oakley, Peter Brocklehurst, Ron Gray National Perinatal Epidemiology Unit, University of Oxford June 2010 Summary Congenital anomalies are the second commonest cause of infant deaths in Engl Wales; although the vast majority of infants born with a congenital anomaly will survive. There are many different congenital anomalies the cause of most is not known. In any single year infant deaths due to congenital anomalies are associated with over 150 different causes. Congenital anomalies contribute about one third of the extra infant deaths experienced by the routine manual socio-economic groups compared with the population as a whole. The primary prevention of congenital anomalies requires delivery of high quality universal pre-pregnancy pregnancy care for women in general together with optimisation of management before conception for specific women at higher risk of anomalies, e.g. diabetics. The availability of genetic services is essential for couples with a family history or past history of pregnancies affected by congenital anomalies, particularly due to specific genetic conditions. Some sections of the population are at increased risk of particular genetic conditions as a consequence infant mortality rates due to congenital anomalies may be higher in some areas compared with Engl Wales overall. Secondary prevention of congenital anomalies is achieved through prenatal screening diagnosis followed by the offer of termination of pregnancies affected by major anomalies. Because termination of pregnancy is the only option for many congenital anomalies secondary prevention is not universally acceptable. Infant mortality rates due to congenital anomalies are strongly influenced by prenatal screening diagnosis, uptake of termination of pregnancy. However, it is important to note that many of the anomalies subject to primary secondary prevention would not in any case have resulted in an infant death thus their prevention, although of great importance because of the associated morbidity disability, will not necessarily impact on the infant mortality rate. 1 Introduction Congenital anomalies are the second commonest cause of infant deaths in the United Kingdom. This briefing paper is the final in a series of four papers commissioned by the Department of Health as part of a project to evaluate the evidence base of interventions to reduce infant mortality inequalities in the infant mortality rate. 1,2,3 The aim of this paper is to provide a background to the evidence review process by examining the contribution of congenital anomalies to the infant mortality rate; inequalities in the rate; the role of prevention; how variations in risk factors uptake of preventive strategies might affect the infant mortality rate associated with congenital anomalies. Definitions of terms used in this paper are given in the glossary at the end. 2 Congenital anomalies 2.1 Definitions Congenital anomalies, congenital abnormalities, birth defects congenital malformations are all terms used to describe developmental disorders of the embryo fetus. There are several hundred separate anomalies which fall under these headings including structural, functional, metabolic hereditary conditions. However, there is no single universally accepted system of classification of anomalies or indeed a single agreed definition of what constitutes a congenital anomaly. For the purposes of this paper we will use the term congenital anomalies take this to mean an embryonic or fetal developmental 1

2 abnormality of any type, but noting that not all variations in development are anomalies as anatomical variation in humans is common. There are four distinct types of anomalies which result from different sets of causes: 4 Malformations are caused by intrinsically abnormal processes during the development of the egg or the sperm, or during fertilisation. These include the anomalies that are caused by chromosomal abnormalities, for example having an extra chromosome 21 in Down s Syndrome, single gene defects, for example, campomelic syndrome. Disruptions refer to defects that are caused by the interference with an originally normal developmental process. Disruptions can be caused by teratogens such as drugs, for example alcohol, thalidomide warfarin; chemicals, for example polychlorinated biphenyls (PCBs); viruses, for example cytomegalovirus (CMV) rubella; ionising radiation, for example X-rays. Deformations are abnormal forms, shapes or positions of a part of the body result from mechanical forces. For example, twins can suffer limb deformation such as clubbed foot due to the crowded uterine space. Dysplasia refers to the abnormal organisation of cells into tissue. The causes are generally nonspecific as a consequence often affect several organs simultaneously. The frequency of occurrence of congenital anomalies is usually quoted as birth prevalence rather than incidence. This is because we know that many fetuses affected by a congenital anomaly will miscarry, that the miscarriage may occur before the pregnancy is confirmed even if the miscarriage is later the anomaly may not be diagnosed. The term birth prevalence therefore acknowledges that not all cases of congenital anomalies can be diagnosed counted in an incidence rate. The denominator for the calculation of birth prevalence is usually taken as all births which excludes miscarriages terminations of pregnancy. 2.2 Sources of information about congenital anomalies in Engl The National Congenital Anomalies System (NCAS) was established in 1964 in the wake of the thalidomide disaster as a surveillance system intended to identify hazards such as thalidomide quickly. The system has operated by issuing monthly alerts to Directors of Public Health when there are changes in the frequency of reported anomalies as a means of highlighting possible clusters of cases i. NCAS was not established as a register intended to ascertain all cases of congenital anomalies in the population; its purpose was to concentrate on identifying changes in the frequency of occurrence of reports of anomalies rather than estimating the birth prevalence of conditions. It is only in those areas where a regional congenital anomaly register supplies data to NCAS that robust birth prevalence estimates can be derived from NCAS data. 5 A further limitation of increasing importance in the context of prenatal screening diagnosis the impact on the infant mortality rate is that NCAS does not collect information about pregnancies affected by a congenital anomaly that are terminated. For conditions with a high rate of prenatal detection termination, for example neural tube defects (NTDs), this leads to a substantial under-ascertainment of cases means that if, for example, primary prevention of NTDs through folate fortification of flour is introduced in the UK, as it has been elsewhere, NCAS will not be able to provide useful data for the evaluation of the effects of fortification. The problems with NCAS data may, however, be of limited concern in the future since the whole of the NCAS system is currently under review (consultation closed March 2010) with the proposal that data collection will cease a final report will be issued in 2009/2010. Whilst a legacy data system will be retained subsequent data collection will cease. Discussions about how to provide surveillance data alerts of potential clusters in the short longer term are in progress with the relevant members of the British Isles Network of Congenital Anomalies Registers (BINOCAR). 6 Congenital Anomalies Registers are highly organised systems for collecting data about congenital anomalies with the purpose of identifying all the cases of congenital anomalies in geographically defined populations. Most Registers in the UK collect information about anomalies diagnosed prenatally, those diagnosed in the newborn early childhood in specific designated geographical regions. 7 In the UK the network of regional congenital anomalies registers is represented by BINOCAR (Figure 1). 6 The regional congenital anomalies registers are now the only source of populationbased congenital anomalies data in the UK able to: support on-going audit evaluation of the national prenatal screening diagnosis programme; provide routine prevalence trend information about anomalies; the only means to systematically identify investigate potential clusters of cases. i NCAS ceased issuing alerts in mid-2009 NCAS has subsequently closed completely. 2 The contribution of congenital anomalies to infant mortality

3 Geographically the BINOCAR registers cover much of the area of Engl, however, mainly because of the concentration of births in the East, London the South East of Engl the lack of registers in those areas, only about 50% of births in Engl are born in areas served by a BINOCAR register. Two of the registers are however, at risk with funding in place only to March 2011: Congenital Anomalies Register for Oxfordshire, Berkshire Buckinghamshire (CAROBB) i the Northern Congenital Anomalies Survey (NorCAS). ii Whereas, a new register is proposed for Yorkshire Humber (YHCAR). Figure 1. BINOCAR Registers Map 6 * BINOCAR Register Proposed BINOCAR Register No BINOCAR Register coverage Cork & Kerry Dublin South Eastern Scotl Glasgow CARIS Wales South West Northern Merseyside & Cheshire Yorkshire & Humber West Midls East Midls & South Yorkshire Oxford, Berkshire & Buckinghamshire Wessex * A BINOCAR register operates in Glasgow; information about congenital anomalies in the rest of Scotl is collected by routine data systems provided by the Information Services Division, NHS Services Scotl 2.3 Frequency of occurrence of congenital anomalies The lack of agreement about the definition of congenital anomalies means that the comparison of the birth prevalence of anomalies is problematic as inclusion exclusion criteria vary between different data sources. Furthermore some registers have historically concentrated on prenatally neonatally diagnosed conditions 10 whereas others in different parts of the world have included anomalies diagnosed from pregnancy into childhood, for example to age 6 years in Western Australia. 11 Variations in birth prevalence estimates may also arise depending upon whether or not figures relating to pregnancies terminated because of fetal congenital anomalies (TOPFA) are included. Variations in methods of ascertainment are also important. Passive systems of case notification tend to identify fewer cases compared with more active ascertainment. 7 As a consequence of all these differences estimates of birth prevalence vary from place to place. However, apart from a small number of specific anomalies, any differences in prevalence are thought largely to be due to methodological differences rather than true differences in underlying population incidence. About 3% of fetuses newborns are diagnosed with a congenital anomaly in the UK each year 10,12 either before or soon after birth; this includes fetuses which are terminated because of the presence of a congenital anomaly (TOPFA). This means that in 2008 when there were 712,328 live births stillbirths in Engl Wales over 21,000 were affected by a congenital anomaly diagnosed before or around birth. 3 Contribution of congenital anomalies to infant deaths 3.1 National statistics Since the introduction of the current stillbirth neonatal death certificate in 1986 it has not been possible to directly compare the causes of neonatal deaths with the causes of postneonatal deaths. 1 This is because, in contrast to deaths in general (including postneonatal deaths), no single underlying cause of death is recorded for neonatal deaths. As a consequence, the Office for National Statistics (ONS) developed a hierarchical classification which allows both neonatal postneonatal deaths to be assigned to a specific category based on the likely timing of the damage leading to the death (see glossary for the definition of the ONS cause of death classification algorithm). The causes of death given in Figure 2 are shown in the hierarchical order of the ONS cause groups. i CAROBB website available at: carobb [Accessed April 2010] ii NorCAS website available at: rmso/surveys/congenital [Accessed April 2010] The contribution of congenital anomalies to infant mortality 3

4 Figure 2. ONS cause group-specific mortality rates by timing of the death*, Engl Wales, Infant mortality rate per 1,000 live births Infant Neonatal Postneonatal Timing of the death 1. Congenital anomalies 2. Antepartum infection 3. Immaturity 4. Intrapartum causes 5. External conditions 6. Infections 7.Other specific causes 9.Sudden infant deaths 0. Unclassified *Neonatal deaths deaths occurring under 4 weeks of age Post-neonatal deaths - deaths occurring from 4 weeks to one year of age Infant deaths deaths occurring before one year of age Congenital anomalies are the second commonest cause of infant deaths overall with a rate in 2007 of 1.39 per 1,000 live births; they are the leading cause of deaths in the postneonatal period at 0.52 per 1,000 live births The proportion of infants with a congenital anomaly who die The vast majority of pregnancies births affected by a congenital anomaly survive do not result in a stillbirth or infant death. In infant deaths 557 stillbirths were ascribed to congenital anomalies. That is, of the 3% of pregnancies infants diagnosed with an anomaly only 7% of them resulted in a stillbirth or infant death; 93% of affected cases survived birth infancy. Furthermore the risk of stillbirth, neonatal death later deaths varies between different anomalies. 3.3 Congenital anomalies which result in an infant death Mortality associated with congenital anomalies varies enormously depending upon the particular anomaly. For example anencephaly (failure of the development of the skull brain) is incompatible with life. The vast majority of fetuses affected by anencephaly are either detected prenatally the pregnancy is terminated, or the baby is stillborn. Figure 3 illustrates the distribution of neonatal postneonatal deaths in Engl Wales 2002 to 2005 where the main cause of death was a congenital anomaly, by the system affected. Figure 3 also shows the distribution of all cases congenital anomalies taken from the East Midls South Yorkshire Congenital Anomalies Register (EMSYCAR). 12 This illustrates the difference in the proportion of actual cases of different types of anomalies compared with the proportion of deaths from the different anomalies. Note that the distributions are shown as proportions of deaths proportions of cases by category of anomaly; there are many more actual cases than deaths. Figure 3. Distribution of causes of neonatal postneonatal deaths from congenital anomalies (2002 to 2005) 13 the distribution of all congenital anomalies cases from the East Midls South Yorkshire Congenital Anomalies Register (EMSYCAR) (births 1997 to 2001) 12 * Percentage 50 All EMSYCAR cases* Circulatory system Respiratory system Musculoskeletal system Nervous system Chromosomal Other anomalies Urogenital tract Postnatal deaths Neonatal deaths Rest digestive system Cleft lip /or palate *Note the figures given are percentages of deaths percentages of cases. There are about 14 times more cases of congenital anomalies than deaths due to anomalies The commonest anomalies affect the musculoskeletal urogenital system, although these are relatively uncommon causes of death. Whereas, the commonest causes of death from anomalies are due to anomalies affecting the circulatory system (mainly the heart) which are only the third commonest cause of anomalies overall. Many different anomalies are responsible for these deaths as illustrated by the detailed neonatal death information provided in routine ONS publications. In 2007 there were 482 neonatal deaths for which the main cause was a congenital anomaly 13 with 165 different congenital anomalies recorded as the main cause of death. Furthermore, the 165 different categories included several which are defined as other which includes a series of separate conditions so that the 165 is an underestimate of the very wide range of different anomalies associated with neonatal mortality. The congenital anomalies which result in an infant death are those which meet the following conditions: 4 The contribution of congenital anomalies to infant mortality

5 They are: Not universally lethal during pregnancy Have a high risk of early death either due to the anomaly itself or the complications of treatment Are NOT detectable (or have a low detection rate) through screening during pregnancy Or Are detectable through prenatal screening but there is poor access to screening, or low uptake of screening, or termination of pregnancy tends not to be taken-up The group of anomalies which best illustrates this typology is the heart anomalies group. There are over 80 individual heart anomalies described in the ICD10 classification system, including several categories of other in which many individual, rarer anomalies may be classified. Unless they form part of a syndrome, heart defects tend not to result in stillbirth. The overall prenatal detection rate for cardiac anomalies is relatively low at about 35-40% 10 even with the sophisticated surgery now available the complex cardiac anomalies generally have a relatively high risk of infant death. 3.4 Variations in infant mortality due to congenital anomalies Variations in infant mortality rates due to congenital anomalies between different groups in the population arise under the two sets of conditions outlined in Figure 4. The typology outlined in Figure 4 makes explicit the points (*) at which interventions might theoretically be put in place to reduce variations between groups in the rate of infant deaths due to congenital anomalies. The two main points for intervention to reduce variation are first the underlying aetiological risk, second in the access to screening. A third potential point is in differential uptake of screening for which there may be variation due to lack of information. However, even assuming that full information is given to all women, variations in choice about prenatal screening between different groups of women are likely to remain. 15 These are very personal choices mainly relating to termination of pregnancy for which there may be strong cultural, moral religious influences, different views of what is regarded as a good reproductive outcome. There is a fourth potential intervention point if there is differential access to treatment once a baby with a congenital anomaly is born. In the context of the NHS any variations in access quality of treatment, where they exist, are likely to be geographical rather than based Figure 4. Conditions under which variations in infant mortality rates due to congenital anomalies might arise between different groups in the population Conditions 1: There are NO variations between groups in the underlying risk of the anomalies; incidence is the same The anomalies are not universally lethal during pregnancy The anomalies have a high risk of early death due to the anomalies or the complications of treatment The anomalies are detectable through prenatal screening BUT there is differential screening access (*), or differential screening uptake, or differential uptake of termination of pregnancy (TOPFA) Conditions 2: There ARE variations between groups in the underlying risk (*) of the anomalies resulting in differences in incidence The anomalies are not universally lethal during pregnancy The anomalies have a high risk of early death due to the anomalies or the complications of treatment o o The anomalies are NOT detectable (or have a low detection rate) through prenatal screening Or The anomalies are detectable through prenatal screening BUT there is differential screening access (*), or differential screening uptake, or uptake of termination of pregnancy (TOPFA) (*) Points at which it may be possible to introduce interventions to reduce variations in infant mortality due to congenital anomalies between different groups in the population The contribution of congenital anomalies to infant mortality 5

6 on individual patient characteristics. However, providing information to parents to enable them to make fully informed treatment choices in relation to complex high risk procedures may be particularly difficult for some groups, for example, where English is not their first language. To intervene on the underlying risk of anomalies (primary prevention) requires knowledge of potentially modifiable risk factors, the means to modify the risk factors the widespread uptake of those interventions. Risk factors for congenital anomalies are discussed later. To intervene on access to prenatal screening (secondary prevention) requires an understing of variations barriers to access. However, even if screening access is universally equal, there is some evidence that this is not the case, 15 there may nevertheless be groups in the population who choose not to take up prenatal screening diagnosis 15 or, where they take up prenatal screening diagnosis they may act on a positive result by choosing to continue with their pregnancy parent their child as an alternative to choosing to terminate the pregnancy. 3.5 Evidence of variations between groups in infant mortality rates due to congenital anomalies The available national data relating to congenital anomaly deaths only allows examination of a small number of different groups in the population. Figure 5 illustrates differences in the rate of infant death by the ethnic group of the mother highlights the statistically significantly four-fold higher risk of infant death from congenital anomalies faced by babies born to mothers of Pakistani origin compared with all the other groups listed. This higher risk represents about 90 extra deaths per year in the infants born to Pakistani mothers in Engl Wales over the number of deaths that would have been expected in this group had they experienced the same infant mortality rate as White British mothers. Babies born to mothers from other ethnic groups are also at a statistically significantly higher risk of infant death compared with infant born to White British mothers of about 45%. This represents an excess of about 25 deaths per year compared with the number that would have been expected had they experienced the same infant mortality rate as White British mothers. Figure 5. Infant mortality rates due to congenital anomalies with 95% confidence intervals, by ethnic group per 1,000 live births for babies born in Engl Wales in Infant mortality rate due to congenital anomalies per 1,000 live births *Statistically significantly higher than all the other groups **Statistically significantly higher than the White British group **Chinese, Other Asian, Other Black, Other all Mixed groups Ethnic group of mother derived from linkage of birth registration statistics to NHS numbers for babies (NN4B) data 1,3,16 Figure 6 illustrates differences in the rate of neonatal death due to congenital anomalies by the socio-economic position of the neonate based on paternal occupation coding. This highlights a statistically significantly doubling in the risk of neonatal death due to congenital anomalies in the groups of babies born to fathers in the NS-SEC categories of 6 7 those whose occupations could not be classified, compared with babies born to fathers in the NS-SEC groups 1-5. This higher risk represents about 88 extra deaths per year in groups 6,7 unclassified over the number of deaths that would have been expected in this group had they experienced the same mortality rate as the infants born to fathers in the NS-SEC groups 1-5. In trying to underst these differences it is important to note that the figures relating to ethnicity socio-economic position are not independent it is likely that some ethnic groups are more likely to have a higher or lower socio-economic position that is, these two factors confound each other. It is not possible to untangle these effects in relation to infant mortality from ONS published data because the data are only available in aggregated form. However, work undertaken by the Bradford Infant Mortality Commission demonstrated that in Bradford 88% of babies in born to Pakistani mothers 41% of babies born to white British mothers were born in the most deprived two-fifths of neighbourhoods. 17 For infant deaths overall a further analysis of the Bradford data was suggestive of an independent excess infant mortality for mothers of Pakistani origin over above the effects of socio- 6 The contribution of congenital anomalies to infant mortality

7 economic position. These results were not statistically significant but this was not surprising given the small numbers of deaths involved. Figure 6. Neonatal mortality rate due to congenital anomalies with 95% confidence intervals, by socio-economic position (NS- SEC) per 1,000 live births for babies born in Engl Wales 2002 to 2005± Neonatal mortality rate due to congenital anomalies per 1,000 live births * 2* * 7* Not class* Socio-economic postion (NS-SEC) Sole reg* For main cause of death ±Derived from a data download provided by ONS *Statistically significantly different than the overall rate Overall 3.6 The contribution of congenital anomalies to the gap in infant mortality between the routine manual groups the population as a whole The aim of the infant mortality public service agreement target for infant mortality is:...to reduce by at least 10% the gap in the mortality between the routine manual groups the population as a whole Using the data provided by ONS for Engl Wales illustrated in Figure 6 we have estimated the excess number of infant deaths per year for the routine manual groups in the population have calculated the contribution of congenital anomalies to the excess. For the period 2002 to 2005 the routine manual groups experienced, on average, 160 extra infant deaths above the number that would have been expected had they had the same infant mortality rate as the Engl Wales population as a whole. Infant deaths due to congenital anomalies contributed about one third of this total that is, 50 of these extra deaths each year. Of note, these are average national population figures are likely to vary across localities due to differences in the population characteristics. 4 Causes prevention of congenital anomalies 4.1 The causes of congenital anomalies The cause of the majority of congenital anomalies is not known. 4 There is a small number of very specific anomalies associated with particular conditions. These include, the teratogenic effects of particular drugs taken during pregnancy, for example, abnormal or absent limbs (phocomelia) associated with thalidomide, valproate embryopathy associated with sodium valproate treatment for epilepsy, characteristic skeletal abnormalities associated with warfarin ingestion during pregnancy. 19 Some infectious diseases also cause characteristics anomalies, for example rubella infection contracted during pregnancy can result in the classic triad of: eye heart anomalies with sensorineural deafness. Some women who have had a previous pregnancy affected by a congenital anomaly are at an increased risk of having a further affected pregnancy, for example the recurrence risk of neural tube defects (spina bifida anencephaly) is about 5%. Some specific anomalies are known to have a genetic origin where particular gene mutations or deletions have been identified. For example, most cases of Apert s syndrome are due to a spontaneous mutation affected one of two genes. The cause of the mutations is not known but when one or other mutation is present Apert s syndrome results. Apert s syndrome has an autosomal dominant inheritance, so that someone with Apert s syndrome has a 1 in 2 chance of passing the condition on to their children. Regardless of the pattern of inheritance, genetic conditions associated with a high infant mortality rate, for example, campomelic syndrome generally occur sporadically because affected infants tend not to survive. However, a couple who has had a child affected by a genetic condition are likely in any subsequent pregnancy to have an increased risk of having a further affected child will require genetic counselling. Depending upon the condition, in subsequent pregnancies prenatal diagnosis through amniocentesis the termination of an affected pregnancy, will be offered if an appropriate genetic test is available. Pre-implantation genetic diagnosis may also be appropriate if couples are willing able to undergo in vitro fertilisation (IVF) treatment. By identifying transferring only unaffected embryos, pre-implantation genetic diagnosis avoids the need for termination of an affected pregnancy. The contribution of congenital anomalies to infant mortality 7

8 Some conditions are associated with a general increase in the risk of groups of congenital anomalies, although not associated with single specific conditions or syndromes. For example, pregestational maternal diabetes is associated with an increase in the risk of congenital anomalies overall 20 there is evidence that gestational diabetes is associated with a similar two fold increase in the risk of anomalies overall. 21 While cardiac, skeletal central nervous system anomalies are the most common, diabetes-associated anomalies usually involve one or more organs. Whilst the exact mechanism is unclear hyperglycaemia is thought to be the primary teratogen causing, in particular, cardiac anomalies. 20 As a consequence the NICE guidance for diabetic women during pregnancy recommends that women with diabetes who are planning to become pregnant should be told that establishing good glycaemic controls before conception continuing this throughout the pregnancy will reduce the risk of miscarriage, congenital malformation, stillbirth neonatal death. 22 Women who are obese are also at an increased risk of a range of congenital anomalies, including cardiovascular anomalies, neural tube defects, cleft palate, hydrocephaly limb reduction defects. 23 Again, the causal mechanism(s) underlying these associations is unclear. Obesity diabetes share common metabolic abnormalities including hyperglycaemia insulin resistance undiagnosed diabetes hyperglycaemia are possible explanations for the increased risk of congenital anomalies in the offspring of obese women. Nevertheless, whilst the evidence is conflicting there does appear to be an independent risk of obesity above that associated with diagnosed diabetes. 23 A role for nutritional deficiencies associated with obesity, including reduced folate levels, has also been postulated. Further generally non-specific risk factors for congenital anomalies include maternal age, cigarette smoking, alcohol drug use. The risks of anomalies in general chromosomal anomalies in particular increase with increasing maternal age; there is also growing evidence of the effects of older paternal age on the risk of dominant gene mutations, for example, Apert s syndrome. There are some specific exceptions to the increase in risk with maternal age, for example the risk of gastroschisis is inversely related to maternal age with the peak birth prevalence in women <25yrs. The risk of gastroschisis is also increased with use of recreational drugs, particularly those with a vasoconstrictive action, for example, cocaine, amphetamines ecstasy. 24 The role of consanguinity as a risk factor for congenital anomalies infant death is complex studies in the past often failed to account for socio-economic circumstances other important confounders. 25 Data from Pakistan, in the early 1990s where 61% of marriages were between first (50%) second cousins (11%), enabled adjustment for socio-economic other factors. Against a background infant mortality rate of 99 per 1,000 live births, infants of first cousins had a 42% increased risk of death in the first year for second cousins the increase was 24%. 25 The overall contribution of consanguinity to infant mortality generally infant mortality due to congenital anomalies in the UK will inevitably be small since first second cousin marriages are generally uncommon. 26 However, as was found in Bradford, the burden will be disproportionally borne by groups in the population where cousin marriages are more common, for example, couples of Pakistani origin (Figure 5). 17 Of note consanguinity is particularly associated with genetic conditions which have an autosomal recessive pattern of inheritance. 4.2 Primary prevention of congenital anomalies The primary prevention of congenital anomalies is only possible for a very small range of specific anomalies for which there is either a known cause, or even in the absence of a clear understing of the cause, a means of prevention has been identified. On a population level these include childhood rubella immunisation, screening treatment for syphilis during pregnancy, periconceptional folic acid supplementation /or folate food fortification for the prevention of neural tube defects. On an individual level, optimising the management of women at higher risk, for example, for women who are diabetic or epileptic, is the ideal approach to minimising the risks of anomalies. However, as about 40% of pregnancies in the UK are unplanned 27 this approach is not always possible even in those women at higher risk of problems in their pregnancy. Furthermore, since relatively few of the anomalies given in these examples would result in deaths in infancy, whilst the goal of reducing the risk of anomalies is important, it may have relatively little impact on the infant mortality rate. 4.3 Secondary prevention of congenital anomalies There are two main methods of prenatal screening relating to congenital anomalies available in the UK. Screening for Down s syndrome (trisomy 21) which affects about 2.6 per 1,000 pregnancies is offered in the first or second trimesters. 28 Screening for Down s 8 The contribution of congenital anomalies to infant mortality

9 may also detect the rather rarer conditions of Edward s syndrome (trisomy 18; 0.7 per 1,000 pregnancies) Patau s syndrome (trisomy 13; 0.2 per 1,000 pregnancies). 28 Although Down s syndrome is associated with an increased risk of miscarriage or stillbirth, it is a rare cause of infant death when a death does occur is it usually due to a cardiac anomaly to which infants with Down s have an increased risk. From maternal report it is estimated that about 65% of women take up the offer of screening for Down s syndrome. 15 Lack of universal uptake probably reflects first, a failure of universal offer of screening the extent to which this occurs the reasons why are unclear; 15 second, a desire by some women not to terminate their pregnancy even if the fetus is affected by Down s syndrome. Perceptions of a good reproductive outcome are very personal influenced by many social, cultural religious factors. The second main form of universal prenatal screening is detailed second trimester ultrasound scanning usually offered between 18 to 20 completed weeks gestation which is designed to assess fetal growth identify structural congenital anomalies including some structural anomalies associated with chromosomal defects. Despite the fact that the consequences of detection of anomalies through this route is the same as for Down s syndrome, uptake of second trimester ultrasound scanning is higher than the uptake of Down s syndrome screening. This is possibly because many women may be unclear about the purpose of the scan view it primarily as a source of reassurance of pictures of their baby. 29 Prenatal detection rates vary by anomaly (Table 1). It is important to note however, that only a relatively small proportion of these cases would contribute to the infant mortality rate even if they remained undetected prenatally. The main contributors to infant deaths for these selected anomalies would be a minority of the cardiac defects, the more severe cases of congenital diaphragmatic hernia a minority of the chromosomal anomalies. Table 1. Prenatal detection rates for selected isolated* congenital anomalies, data from the Congenital Anomalies Register for Oxfordshire, Berkshire Buckinghamshire (CAROBB) Selected isolated* anomalies Detection rate (%) 95%CI Neural tube defects 95% 89%-98% Cardiac anomalies 35% 30%-42% Cleft lip +/- palate 65% 53%-75% Congenital diaphragmatic hernia 64% 41%-83% Gastroschisis 100% 93%-100% Exomphalos 90% 68%-99% Chromosomal 56% 50%-62% *Isolated no other congenital anomalies present On an individual basis couples identified as being at increased risk of genetic conditions will be offered genetic counselling where suitable tests are available, they will be offered prenatal diagnosis through amniocentesis in a subsequent pregnancy. As discussed above pre-implantation genetic diagnosis may also be an option. 5 Conclusions Congenital anomalies are the second commonest cause of infant deaths in Engl Wales, although the vast majority of infants born with a congenital anomaly will survive with a near normal life expectancy. The underlying cause of most congenital anomalies is not known. There are many different congenital anomalies in any single year infant deaths due to congenital anomalies in Engl Wales are associated with over 150 different causes. Congenital anomalies contribute about one third of the extra infant deaths experienced by the routine manual groups compared with the population as a whole. The primary prevention of congenital anomalies requires many different activities the delivery of a high stard of universal pre-pregnancy pregnancy care for women in general together with optimisation of management before conception for specific women at higher risk of anomalies, for example, women with diabetes, epilepsy those taking other specific drugs with teratogenic effects. The availability of genetic services is essential for couples with a family history or past history of pregnancies affected by congenital anomalies, particularly due to specific genetic conditions. Some sections of the population are at increased risk of particular genetic conditions as a consequence infant The contribution of congenital anomalies to infant mortality 9

10 mortality rates due to congenital anomalies may be higher than in the population in general; genetic services will be of particular importance in these areas. Secondary prevention of congenital anomalies is achieved through prenatal screening diagnosis followed by the offer of termination of pregnancies affected by major anomalies. Because termination of pregnancy is the only therapeutic option for many congenital anomalies secondary prevention is not universally acceptable. Infant mortality rates due to congenital anomalies are strongly influenced by prenatal screening diagnosis, uptake of termination of pregnancy. However, it is important to note, that many of the anomalies which can be prevented through primary or secondary actions would not in any case have resulted in an infant death thus their prevention, although of great importance because of the associated morbidity disability, will not necessarily impact on the infant mortality rate. Glossary Apert s syndrome a rare congenital disorder characterised by malformations of the skull, face, hs feet which requires extensive surgery. It is almost always caused by a spontaneous gene mutation of paternal origin the risk of Apert s syndrome increases with the age of the father. Autosomal dominant genetic inheritance this is the mechanism of inheritance of genetic conditions which are caused by a single gene inherited from one parent. Someone who carries the gene for an autosomal dominant condition will always be affected by the condition each of their children will have a 1 in 2 chance of inheriting the condition. Autosomal recessive genetic inheritance this is the mechanism of inheritance of genetic conditions which requires a single gene from both parents to be affected in order for their child to be affected. Carriers of recessive genes are themselves not affected but if they have a child with someone who is also a carrier each child they have will have a 1 in 4 chance of being affected by the condition. BINOCAR British Isles Network of Congenital Anomalies Registers. Birth prevalence is the measure of frequency of occurrence usually used in relation to congenital anomalies. The term incidence is not used in acknowledgement of the fact that many embryos fetuses affected by a congenital anomaly are miscarried therefore not counted in the rate calculation. Birth prevalence is the number of new cases of congenital anomalies born (live births stillbirths) divided by the total number of births. Campomelic syndrome a rare single gene defect with autosomal dominant inheritance which results in abnormalities of the bones the cartilage of the respiratory tract. Almost without exception affected babies die in the neonatal period from respiratory complications. The characteristic bone features can be diagnosed on prenatal ultrasound. There is no treatment available, other than to provide respiratory support to babies born alive. It affects about 1 in 500,000 pregnancies. CAROBB Congenital Anomalies Register for Oxfordshire, Berkshire Buckinghamshire. Consanguinity unions contracted between persons biologically related as second cousins or closer. This arbitrary limit has been chosen because the genetic influence in offspring from marriages between couples related to a lesser degree would usually be expected to differ only slightly from that seen in the general population. Globally, the most common form of consanguineous marriage is between first cousins, in which the spouses share 1/8 of their genes inherited from a common ancestor. Down s syndrome is one of the more common chromosomal disorders due to having part or all of an extra chromosome number 21. It is characterised by a series of major minor structural abnormalities, for example, a small chin, an unusually round face, a large tongue almond shaped, widely spaced eyes. Cognitive impairment is usually present, although the extent is variable. Health concerns include a higher risk of congenital heart anomalies, recurrent ear infections, obstructive sleep apnoea, thyroid disease, leukaemia early onset Alzheimer s disease. Edward s syndrome is a chromosomal disorder due to having part or all of an extra chromosome number 18. Infants have multiple anomalies affecting the heart, kidney, intestines (exomphalos). The majority of affected infants die within the first month after birth only 10% survive to age one year. Long term survival is very uncommon. Embryo In humans during pregnancy, an embryo is the developing organism from the time of fertilisation until the end of the eighth week of gestation, following which it is called a fetus. 10 The contribution of congenital anomalies to infant mortality

11 Exomphalos is a defect in the abdominal wall through the umbilicus through which the intestines other organs develop outside the abdomen. It is more commonly associated with other congenital anomalies, for example trisomy 18, than gastroschisis as a consequence the long term outcome is less favourable than for gastroschisis; termination of affected pregnancies is more common. Fetus In humans during pregnancy, a fetus is the developing organism from the end of the eighth week of gestation until birth. Gastroschisis is a defect in the abdominal wall to one side of the umbilicus through which the intestines other organs develop outside of the abdomen. It is rarely associated with other congenital anomalies. Surgical repair of the defect is required following birth with the majority of the infants treated having a good long-term outcome although complications can arise. ICD10 classification system The International Statistical Classification of Diseases Related Health Problems 10th Revision (ICD-10) is a coding of diseases signs, symptoms, abnormal findings, complaints, social circumstances external causes of injury or diseases, as classified by the World Health Organisation. Incidence rate is a measure of the occurrence of new cases of a disease in a population. This is in contrast to prevalence which is a measure of both new cases existing cases in the population. In the field of congenital anomalies the term birth prevalence is rather confusingly used to describe new cases of anomalies. Birth prevalence is used because we know that many new cases of anomalies do not survive in pregnancy long enough to be identified since they are miscarried. We therefore cannot know about all new cases of a particular anomaly. Infant mortality rate number of deaths at ages under one year per 1,000 live births. Live birth birth of an infant showing any signs of life regardless of gestation at birth. National Statistics Socio-Economic Classification (NS-SEC) a method of coding socio-economic position on the basis of occupation introduced in 2002 to replace the Stard Occupational Classification which coded occupation to the Registrar General s Social Class. NS-SEC has a series of analytical classes which are nominal in the extended 14 operational categories form which become ordinal when collapsed into the three-class version. Infant mortality statistics are published using the eight-class analytical version: 1.1 Large employers higher managerial 1.2 Higher professional 2 Lower managerial professional 3 Intermediate 4 Small employer own-account workers 5 Lower supervisory technical 6 Semi-routine 7 Routine 8 Never worked long-term unemployed Unclassifiable NCAS National Congenital Anomalies System for Engl Wales. Neonatal mortality rate number of deaths at ages under four weeks per 1,000 live births. NorCAS Nothern Congenital Abnormality Survey. NS-SEC see National Statistics Socio-Economic Classification above. Neural tube defects a group of conditions caused by the failure of closure of the neural tube during embryonic development during the first 28 days after fertilisation. The neural tube forms the brain spine the type of neural tube defect which occurs depends upon where on the spine or brain the failure to close occurs. If it affects the brain a condition called anencephaly results where the brain fails to develop; a condition incompatible with life. When it affects the spine, spina bifida results its effects depend upon where on the spine the defect occurs; the higher up the more serious the effect the more complications that result. ONS cause (of death) groups hierarchical classification a classification system developed to enable comparison of neonatal postneonatal deaths following a change in deaths certificates for neonatal deaths stillbirths whereby a single underlying cause of death was no longer assigned. This system allows the death to be assigned hierarchically to a specific category based on the likely timing of the damage leading to the death: Before the onset of labour 11. Congenital anomalies 12. Antepartum infections 13. Immaturity related conditions In or shortly after labour 14. Asphyxia, anoxia or trauma Postnatal 15. External conditions 16. Infections 17. Other specific conditions The contribution of congenital anomalies to infant mortality 11

12 9. Sudden infant deaths Unclassified 0. Other conditions Patau s syndrome is a chromosomal disorder due to having part or all of an extra chromosome number 13. Infants have multiple anomalies affecting the heart, kidney, other organs including neurological effects. Stillbirth is common the majority of affected infants die within the first month after birth. Postneonatal mortality rate number of deaths at 28 days over but under one year per 1,000 live births Pre-implantation genetic diagnosis requires in vitro fertilisation (IVF) following which a single cell is removed from the developing embryos to test for specific genetic conditions. Only unaffected embryos are then transferred to the uterus. It is only available for genetic conditions where a diagnostic test can be performed. Primary prevention actions taken to avoid disease or injury occurring in the first place. Examples of primary prevention include vaccination against infectious diseases. Secondary prevention actions taken to identify treat an illness or injury early on in its development with the aim of stopping or reversing the problem. Examples include screening for diseases which can then be treated earlier than would usually be the case. Stillbirth birth of an infant at 24 weeks gestation showing no signs of life. Teratogen any agent which causes abnormalities of embryonic or fetal development, examples include specific drugs (eg thalidomide), X-rays some viruses (eg rubella). TOPFA termination of pregnancy for fetal anomaly. TOPFA is offered following prenatal diagnosis of major congenital anomalies. Acknowledgements We would like to thank members of the Inequalities in Infant Mortality Advisory Group for their helpful comments on an earlier draft. Particular thanks to Nirupa Dattani at the Office for National Statistics for providing a data download of mortality rates due to congenital anomalies by socio-economic position. This briefing paper reports on an independent study which was funded by the Policy Research Programme in the Department of Health. The views expressed are not necessarily those of the Department. We are grateful to Dr Judith Budd at the East Midls South Yorkshire Congenital Anomalies Register, who kindly provided the map of the BINOCAR registers. References 1. Kurinczuk JJ, Hollowell J, Brocklehurst P, Gray R. Inequalities in Infant Mortality Project Briefing Paper 1.Infant mortality: overview context. Oxford: National Perinatal Epidemiology Unit, University of Oxford Gray R, Hollowell J, Brocklehurst P, Graham H, Kurinczuk JJ. Inequalities in Infant Mortality Project Briefing paper 2. Health inequalities infant mortality target: technical background. Oxford: National Perinatal Epidemiology Unit, University of Oxford Gray R, Headley J, Oakley L, Kurinczuk JJ, Brocklehurst P, Hollowell J. Inequalities in Infant Mortality Project Briefing paper 3. Towards an understing of variations in infant mortality between different ethnics groups in Engl Wales. Oxford: National Perinatal Epidemiology Unit, University of Oxford Moore KL, Persaud TVN. Chapter 8. Human Birth Defects. In: The Developing Human: Clinically Orientated Embryology. Philadelphia: Saunders p Boyd PA, Armstrong B, Dolk H, Botting B, Pattenden S, Abramsky L et al. Congenital anomaly surveillance in Engl ascertainment deficiencies in the national system. Br Med J 2005;330:27 6. BINOCAR website available at: binocar.org/ [Accessed April 2010] 7. Newton J, Garner S. Disease Registers in Engl. Oxford: Institute of Health Sciences, University of Oxford ISBN CAROBB website available at: ox.ac.uk/carobb [Accessed April 2010] 9. NorCAS website available at: org.uk/rmso/surveys/congenital [Accessed April 2010] 10. Boyd PA, Rounding C, Kurinczuk JJ. Second Report of the Congenital Anomaly Register for Oxfordshire, Berkshire Buckinghamshire (CAROBB) Births 2005 to Oxford: National Perinatal Epidemiology Unit, University of Oxford Bower C, Rudy E, Callaghan A, Quick J, Cosgrove P, Nassar N. Report of the Birth Defects Registry of Western Australia, Number 16: Subiaco, Perth: King Edward Memorial Hospital Kurinczuk JJ, Draper ES, Budd JLS, Perkins MJ, Clarke M. Trent Congenital Anomalies Register. The First Five Years: Congenital Anomalies in Births Leicester: Department of Health Sciences, University of Leicester The contribution of congenital anomalies to infant mortality

13 13. Office for National Statistics. Mortality Statistics, Childhood, Infant Perinatal. Review of the National Statistician on deaths in Engl Wales, Series DH3 number 40. London: Office for National Statistics Tennant PW, Pearce MS, Bythell M, Rankin J. 20-year survival of children born with congenital anomalies: a population-based study. Lancet 2010;375: Rowe R, Puddicombe D, Hockley C, Redshaw M. Offer uptake of prenatal screening for Down syndrome in women from different social ethnic backgrounds. Prenat Diag 2008;28: Moser K, Stanfield KM, Leon DA. Birthweight gestational age by ethnic group, Engl Wales 2005: introducing new data on births. Health Statistics Quarterly Autumn 2008; 39: Helen Brown the Bradford Health Informatics Service team. Bradford Infant Mortality Commission: Analysis interpretation of infant mortality data. Bradford Available at: TheReport.aspx [Accessed April 2010]. 18. Health Inequalities Unit, Department of Health. Review of the Health Inequalities Infant Mortality PSA Target. London: Department of Health Holmes LB. Teratogen-induced limb defects. Am J Med Genet 2002; 112: Corrigan N, Brazil DP, McAuliffe F. Fetal cardiac effects of maternal hyperglycemia during pregnancy. Birth Defects Res A Clin Mol Teratol 2009;85: National Collaborating Centre for Women s Children s Health. NICE guideline: Diabetes in pregnancy. London: Royal College of Obstetricians Gynacologists Stothard KJ, Tennant PW, Bell R, Rankin J. Maternal overweight obesity the risk of congenital anomalies: a systematic review meta-analysis. JAMA 2009;301: Draper ES, Rankin J, Tonks AM, Abrams KR, Field DJ, Clarke M, Kurinczuk JJ. Recreational drug use: a major risk factor for gastroschisis? Am J Epidemiol 2008;167: Grant JC, Bittles AH. The comparative role of consanguinity in infant childhood mortality in Pakistan. Ann Hum Genet 1997;61: prevalence [Accessed April 2010] 27. Jayaweera H, Joshi H, Macfarlane A, Hawkes D, Butler N, Pregnancy Child Birth. In: Dex S & Joshi H (Eds) Children of the 21st Century: From Birth to 9 months. The UK Millennium Cohort Study Series. Policy Press: Bristol Morris JK, De Souza E. The National Down Syndrome Cytogenetic Register for Engl Wales: 2008/09 annual report. London: Wolfson Institute of Preventive Medicine, Queen March University of London Garcia J, Bricker L, Henderson J, Martin MA, Mugford M, Nielson J, Roberts T. Women s view of pregnancy ultrasound: a systematic review. Birth 2002;29: Macintosh MCM, Fleming KM, Bailey JA, Doyle P, Modder J, Acolet D et al. Perinatal mortality congenital anomalies in babies of women with type 1 or type 2 diabetes in Engl, Wales Northern Irel: population based study. Br Med J 2006;333:177 Please cite this document as: Jennifer J Kurinczuk, Jennifer Hollowell, Patricia A Boyd, Laura Oakley, Peter Brocklehurst, Ron Gray. Inequalities in infant mortality project briefing paper 4. The contribution of congenital anomalies to infant mortality. Oxford: National Perinatal Epidemiology Unit, 2010 This briefing paper other Inequalities in Infant Mortality Project reports are available at: The contribution of congenital anomalies to infant mortality 13